Keeper's Bench · Case-Log

Why We Don't Put Heart Monitors on the Crew

Torque & Shield — how a hacker thinks, how an engineer thinks, and how to combine the two.

The Clever Idea

A reader brought us a genuinely sharp proposal. It deserves to be taken seriously, so let's do that first.

The pitch, essentially: insider threat is the hard problem — the intrusion your perimeter can't see because the actor is already inside, wearing a badge, drawing a salary. Detection tooling watches behavior, but behavior is lagging and noisy. So why not go physiological? Put heart-rate sensors on the crew — smartwatch-grade, cheap, continuous. Watch for the spikes. A sysadmin whose pulse jumps while they're touching production at 2 a.m., an accountant's heart racing during an anomalous export — catch the body leaking what the logs won't. Biometrics as insider-threat telemetry.

Credit where due: that is thinking like a hacker. It attacks the weakest assumption in the whole defensive stack — that our instruments see the actual human — and it uses commodity hardware to do it. If the field were ready, this would be a product.

The problem is that the field is not ready, and the autopsy is instructive. So instructive that we think it's worth publishing — with the idea credited — because understanding why this doesn't work teaches more about security than the half-dozen controls you'd bolt on instead.

Pull up two chairs. This takes a red team and a lawyer.

Autopsy One: The Red Team Says Your Sensor Reads the Wrong Variable

The red-team objection starts with the instrument itself.

Galvanic skin response and heart-rate variability — the signals a cheap wearable can capture — measure arousal. Arousal is not guilt, and crucially, it isn't even valence. It does not distinguish the physiological signature of malice from the signature of a man who just got chewed out by his manager, played a tense round of a shooter on his lunch break, or realized mid-SSH-session that he has left the stove on. Fear spikes the numbers. So do anger, joy, a deadline, a coffee, and a phone call from a teenager's school. You cannot red-team an emotion, but you can red-team a detector: an adversary with any tradecraft learns quickly which stress markers they can mask with beta-blockers or mimic by stepping away from the desk — while your honest operator gets flagged for a panic attack.

And the red team knows the lineage of this idea, because it is not new. Strip the modern coating and this sensor stack is a polygraph — continuous, involuntary, and worn all day. The polygraph's scientific standing, after a century of courtroom and intelligence use, is famously poor: arousal correlates with everything and confesses nothing. Taking the least reliable instrument in the deception-detection menagerie and deploying it fleet-wide, involuntarily, and in the background doesn't fix the instrument. It industrializes the error.

Defenders fall for this class of idea for a reason worth naming: the signal feels more real because it comes from the body. Harder numbers, deeper honesty. But data intimacy is not data validity. A bad measurement collected continuously and invasively is worse than a coarse one collected honestly — because it generates confident false positives, and confidence is what makes them dangerous.

Autopsy Two: The Lawyer Says Bring a Change of Clothes

The compliance autopsy is where the idea stops being merely wrong and starts being expensive.

The ADA / medical examination problem. The EEOC addressed this directly in its fact sheet, "Wearables in the Workplace: The Use of Wearables and Other Monitoring Technology Under Federal Employment Discrimination Laws." Per that guidance, an employer that requires wearables collecting information about an employee's physical or mental condition may be conducting a medical examination under the Americans with Disabilities Act — and heart rate is front and center in the EEOC's own examples of what wearables track. If the technology requires employees to provide health information, the employer may also be making a disability-related inquiry, which under the ADA must be job-related and consistent with business necessity. And the EEOC's worked example is a warning shot for our reader's clever pitch: an employer who uses heart rate, fatigue level, or temperature data to infer that an employee is pregnant — and then acts on the inference — can violate federal anti-discrimination law outright. Now imagine what a continuous heart-rate feed lets an ambitious middle manager infer, wrongly, about arrhythmia, anxiety, or medication side effects. Requiring or pressuring employees into continuous biometric monitoring invites the charge that you are running an unauthorized, permanent, whole-body medical screening operation. That is a category of lawsuit with a well-worn path to the courthouse.

The BIPA problem — the state with the sharpest teeth, and the cautionary tale it just lived through. Illinois' Biometric Information Privacy Act carries statutory damages of $1,000–$5,000 per violation, and for years Illinois courts interpreted "violation" the terrifying way: in Cothron v. White Castle System, Inc. (Ill. 2023), the Illinois Supreme Court held that a fresh claim accrues under BIPA with every scan or transmission of biometric data without informed consent — even while the court itself acknowledged, in its own words, the risk of "annihilative liability," and invited the legislature to revisit the statute. In Cothron that meant a fast-food chain faced exposure reckoned per fingerprint scan, per shift, per employee, for years — arithmetic that ultimately produced a settlement reported north of $9 million. That is what happens when a heart-rate program meets a per-reading statute.

Here's the twist your clever idea should study: the Illinois General Assembly took its Supreme Court up on the invitation. Effective August 2, 2024, SB 2979 amended BIPA to provide that repeatedly collecting the same biometric data from the same person by the same method constitutes a single violation — and the Seventh Circuit has since held that the amendment applies retroactively to pending cases. So yes, Illinois blinked. But blink at what the blink means: the state with the strictest biometric privacy statute in the union discovered what its own law did when scaled across a workforce, and rewired it mid-collapse. Even the reformed regime still puts every employee with a wristband on the claim ledger — one to five thousand dollars per person, times headcount, before your counsel has even read the complaint. And the forward-looking part: the Illinois House has considered HB 3024, which would extend BIPA to cover EKG results from wearable devices, such as smartwatches — the exact signal class the heart-monitor fleet would be collecting. Illinois is not done tightening this terrain; it is catching its breath.

Run the arithmetic the red team would: a heart-monitor program is a standing, self-documenting inventory of ADA medical-examination exposure and per-person BIPA claims, refreshed daily by the very sensors doing the monitoring. The reader's clever sensor mesh isn't a detection program — it is an evidentiary machine pointed at its owner.

And that is before we note the thing lawyers say quietly: consent forms from employees who fear for their jobs are not consent. Under the ADA, "voluntary" participation under employer pressure is analyzed as coercion. The program dies even if every wrist signed.

The Schneier Lens

Bruce Schneier has spent a career on the gap this idea falls into. In "The Psychology of Security" (2008) he opened with a line that deserves to be carved over every SOC: "Security is both a feeling and a reality, and they're not the same." The heart-monitor pitch sells a feeling of security — biometric omniscience, bodies confessing, nothing slipping past — while moving the reality of your risk posture nowhere, or backward: new liability surface, new false-positive workload, new reasons for your best people to quit, and no demonstrated reduction in insider events.

Schneier's later work ("A Hacker's Mind") extends the point: any rule or measurement becomes a substrate to be hacked. Build detection on arousal and adversaries will hack arousal — mask it, borrow it, weaponize it to frame a colleague. And Schneier's Law holds here with grim elegance: the reader who thinks of this clever scheme will not be the last person to think of it — which means the countermeasure you're building to detect them will be studied, understood, and used against you by precisely the insiders you fear.

The Boring Alternative That Wins

So what do you actually do about insider threat? The answer is unglamorous and it works.

Behavioral UEBA with consent and transparency. Not the body — the work. Baseline-normal for accounts and access patterns, deviation alerts surfaced to a human analyst, data-handling policies a person can actually read. What does an engineer pull, copy, or export, from where, at what hour, against what history? Behavior is legible, consentable, auditable, and when it misfires it produces a false positive you can explain and correct — unlike an algorithm that has diagnosed your DBA with a racing heart.

Add the practices that make it honest rather than creepy:

None of that sounds like a keynote demo. It also never appears in a courtroom exhibit.

Closing the Loop With the Reader

One more time, because it matters: the heart-monitor pitch was a good idea. It found the real weakness in insider-threat programs — the invisible human — and reached for a new instrument class to close the gap. That's exactly the instinct this site exists to train. Thinking like a hacker, applied to the defense: the move is right, the era is wrong.

Every era of security tools goes through this cycle — a seductive new signal, a red-team autopsy, a compliance reckoning, and then the boring control that quietly does the work. The trick our trade teaches is to run the autopsies before the procurement order. Your clever idea already did the most valuable thing an idea can do: it died cheap, on paper, and in public, teaching everyone who reads this what the next vendor pitch will look like when it arrives wearing a lanyard.

Keep bringing us ideas. We'll keep burying them properly. Gift for gift — the only way this bench works.

Torque & Shield publishes craft, not secrets: patterns and habits you can carry to your own bench, with nothing here tying to any employer, platform, or client. Case-log rubric, Keeper's Bench.

On the record — sources checked for this piece: